RTS experiences?

Post questions here relative to DataStage Server Edition for such areas as Server job design, DS Basic, Routines, Job Sequences, etc.

Moderators: chulett, rschirm, roy

Post Reply
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

RTS experiences?

Post by chulett »

There's been some very interesting discussions recently about Real Time versus Batch processing and the paradigm shift that goes along with moving into that environment.

I feel that the vast majority of work that people like "us" will be doing in the near future will still be of the traditional, batch oriented work we've been doing for awhile now. However, with the advent of 7.0 and RTS (it is RTS -Real Time Services- isn't it?) some folks here may be moving into that world. I was curious...

Are they any folks out there actually doing this now? Would you care to share your situations and experiences with those of us who haven't ventured to the other side of the looking glass yet? I for one would be interested to know what kind of applications people are actually building or preparing to build to take advantage of this new functionality, your trials and tribulations or anything else you'd be willing to share.

Thanks!
-craig

"You can never have too many knives" -- Logan Nine Fingers
Rahul
Participant
Posts: 19
Joined: Wed Oct 29, 2003 10:21 pm

Post by Rahul »

This is a very good topic and i hope people would pour in more on what they have experienced so far on RTI. Has anybody worked on applications integrating MQSeries/EJB with DS. If yes, Can you please let us know what the architecture has been,functionality and what were the benefits out of the new architecture.

Thank you,

Rahul
1stpoint
Participant
Posts: 165
Joined: Thu Nov 13, 2003 2:10 pm
Contact:

RTS

Post by 1stpoint »

I was at a recent Kimball seminar and there was some discussion on real time updates to an existing data warehouse and how to incorporate it into a batch environment.

There are several approaches, probably the simplest would be to incorporate "real time" dimensions that are created on the fly by the real time update process. The obvious problem is fact changes and reapplying changed facts when the batch (non realtime) process runs. One way to handle this is to partition the realtime data separately.
Post Reply